Download Relation of Adult Delinquency, Childhood Victimization and PTSD in

Ben-David. S. and Goldberg, I. (2008). The Relationship of Childhood Victimization, Drug
Abuse and PTSD to Adult Delinquency in a Prison Population. In: N. Ronel, K. Jaishankar and
M. Bensimon (eds.): Trends and Issues in Victimology. (pp. 205-221) Newcastle: Cambridge
Scholar Publishing
The Relationship between Adult Delinquency, Childhood
Victimization, Drug Abuse and PTSD in Prison Population
ABSTRACT
The present investigation directs attention to the role played by PTSD symptoms and
drug abuse in the correlation between childhood victimization and adult criminal
behavior. Using Amos (Analysis of Moment Structure) it was found that whereas drug
abuse indicated non-specific connections between past (childhood/adolescent) types
of victimization and types of adult criminal behavior, PTSD symptoms, such as
intrusive thoughts and cognitive avoidance, led to specific types of criminal behavior
(the same as that of the victimization type). Thus, sexual victims who develop
cognitive avoidance tend to become sexual offenders as adults, while those who
develop drug dependency tend to exhibit non-specific criminal behavior.
Introduction
The purpose of this paper is to introduce a model explaining the ‘cycle of
violence’ that represents the relationship between adult criminal activity and
childhood victimization. The model introduced in this paper suggests that past
victimization, the developmental phase of PTSD and drug usage may serve as an
explanatory factor for the various types of adult criminal activity.
The similarity between childhood/adolescent victimization and certain types of
crime may be explained by PTSD; while PTSD followed by drug abuse may explain e
adult criminal activities that are not related to childhood victimization.
Childhood sexual and physical abuse has been found to increase the probability of
adult violent and criminal behavior. Curtis (1963) maintained that victimized children
are “tomorrow’s murderers and perpetrators of other crimes of violence, if they
survive” (p.386). More recent research results further support this notion (SpatzWidom, 2000, White & Spatz-Widom, 2003).
1
Moreover, childhood or adolescent victimization was shown to have serious short
and long term effects on its victims (Bergen, Martin, Richardson, Allison, Roeger,
2004; Ben-David, Alek & Silfen, 2002). Symptoms such as depression, anxiety,
sleeplessness and intrusive thoughts were so common among abuse victims that, in
1980, Post-traumatic Stress Disorder (PTSD) was defined and introduced in the
Diagnostic and Statistical Manual of Mental Disorder (3rd edition: DSM-III) (Rowan
& Foy, 1993). The aim of this paper is to explore the correlations between childhood
victimization, PTSD, drug abuse and adult criminal behavior.
Various studies have shown that children who were exposed to childhood
victimization tend to be antisocial, aggressive and commit crimes as adults (Bergen et
al, 2004; Saptz-Widom, 2000; Haapasalo & Pokela, 1999). Both the prospective
(Maxfield & Widom, 1996) and retrospective approaches (Clarke, Stein, Sobota,
Marisi, & Hanna, 1999) suggest a high prevalence of abused children in both the
female and male adult criminal populations. This phenomenon is known as the ‘cycle
of violence’ (Spatz- Widom, 1992; Maxfield & Widom 1996).
However, the research on the link between earlier victimization and adult
delinquency has yielded, overall, contradictory results. These results may be divided
into two broad categories (Cohen, 1995): A non-specific General Link (GL), which
means that any form of abuse may lead to various forms of adult violence and/or
criminal behavior (Barnard, Hankins, & Robbins, 1992; Graham, 1996); A Direct
Link (DL), which maintains that the type of victimization predicts specific ensuing
criminal activity, i.e.: sexually abused children will become sex offenders, battered
children will become violent criminals etc. (Laws & Marshall, 1990; Marshall &
Barbaree, 1990).
Several theoretical approaches may explain the connection between past
victimization and adult criminal activity. A few of them include: the social learning
(Bandura, 1973, 1977) and “Monkey see monkey do” hypothesis of Curtis (1963), the
traumagenic hypothesis of Finkelhor (1988), and the attachment insecurity
supposition (Smallbone & MacCabe, 2003). Haapasalo and Pokela (1999) attested
that the trauma model is the most interesting theoretical approach to explain the
relationship between past victimization and present criminal activity. While these
theoretical explanations may account for the SL, none of these theories have tried to
account for both types of links, the General and the Specific.
2
This paper suggests that PTSD and consequential drug abuse may explain these
two links between past victimization and criminal activities.
The DSM-IV divides PTSD into three symptomatic groups: persistent
psychological re-experiencing of the traumatic event, psychological numbing or
avoidance, and hyper arousal (American Psychiatric Association, 1994). However
some researchers have suggested a different classification of PTSD. Based both on
statistical analysis, such as factor analysis, and theoretical considerations, they suggest
that avoidance and numbing are distinct phenomena. Thus, they should be separated
and renamed as cognitive avoidance and emotional avoidance. While cognitive
avoidance is largely regulated by a cognitive psychological process, emotional
avoidance is also mediated by biological-hormonal mechanisms (Foa, Riggs, &
Gershuny, 1995; Myslobodsky, Glicksohn, Singer, Stern, Bar Ziv, Friedland, and
Bleich, 1995).
Everly (1995) suggested a neuro-cognitive model for the developmental sequence
of PTSD. In this model, post trauma, intrusive thoughts are developed first and are
followed by cognitive avoidance. When cognitive avoidance proves ineffective in
reducing intrusive thoughts, the occurrence of emotional avoidance can be expected
(Everly, 1995).
Cognitive avoidance, or a release from intrusive thoughts, can be achieved
through distraction from these painful thoughts. While Emotional avoidance, or the
anesthesia of intrusive thoughts, may be attained by a numbing or blunting of the
emotions (Lipton, 1988). Substance abuse (e.g. alcohol, drugs, food, etc.) may serve
as ‘self-medication’ to help alleviate intrusive thoughts (Teusch, 2001; Saladin,
Drobes, Coffey, Dansky, Brady, Kilpatrick, 2003). Distraction may be achieved
through criminal activity, as suggested by Stott (1950) and Addad (1989).
Although the prevalence of PTSD among both male and female offenders is very
high (Barnard et al, 1992, Yoshinaga, Kadomoto, Onati, Sasaki & Kato, 2004),
limited research examining the correlation between PTSD, criminal activity and drug
abuse exists (Zhang, Welte & Wieczorek, 2001).
Previous studies of Israeli female prisoners found that 71% suffered from past
victimization. A high proportion of these prisoners were characterized by PTSD
avoidance symptoms (Ben-David, 1994; Ben-David, Alek & Silfen 2002). Similarly, a
high frequency of avoidance and hyper-arousal symptoms were found in sex offender
population (Barnard et al, 1992). These findings show the distinction between
3
criminal populations and the intrusive thoughts predominately found in non-criminal
populations (Horowitz, Weine, Jekel, 1995; Solomon, 1988). Thus, it seems plausible
to assume that avoidance, whether cognitive or emotional, is one characteristic of
adult offenders who were previously victimized. It may also be deduced that
avoidance precedes, or is one of the enabling conditions for adults who were
victimized in their youth, the execution of criminal activities.
Criminal activity, similar to any other behavior, is learned. It can also be seen as
an almost natural development of learned behavior, or identification with the
aggressor (Freud, 1966). Abused children learn this behavior in order to reduce stress
or to express disappointment. In addition, children who experience physical violence
or sexual abuse might believe that this is common or acceptable behavior (Finkelhor,
1988). For these children, the normal response to crisis situations in adulthood may be
learned violent behavior similar to the violent or sexual abuse they suffered from in
their youth. These activities are facilitated by emotional avoidance that enables pain to
be inflicted on others. Thus, emotional avoidance and learning theories may explain
the SL.
On the other hand, as aforementioned, drug abuse is one way of releasing
emotional pain after trauma. However, drug abuse may also precede criminal
activities. These activities could manifest as a way to finance the illicit drugs. It
could also be considered indicative of a criminal lifestyle (Bennett & Holloway, 2005;
Brochu, 2001). As a result, victims using drugs as a means to alleviate their former
childhood victimizations are involved in general crimes necessary to finance their
crude remedy. Thus, they are characterized as the GL.
We propose therefore that developmental phase of PTSD (Everly, 1995) may
serve as an explanatory factor for the SL between childhood/adolescent victimization
and the type of criminal behavior manifested in adulthood, while drug abuse following
PTSD may explain the GL.
Method
Participants
The participants included 123 offenders incarcerated in Israeli prisons. The 47
sex offenders (SO) were randomly selected from a list provided by the Israeli Prison
Service. The 76 randomly selected non sex offenders were divided into two groups:
29 violent offenders (VO) and 47 sentenced for miscellaneous offences (MO). The
4
prisoners were assigned to these groups based on their known criminal history,
including the offense for which they had been incarcerated (Cohen, 1995; Nachshon,
1988). The purpose of the research was explained to the participants, and each one
signed a consent form.
Research tools
The data were gathered through two questionnaires:
1. Criminal activity, drug abuse and past victimization inventory.
A Self-disclosure inventory of past victimization, criminal activity and
substance abuse was developed, validated and used in research with female (BenDavid, 1991), and male (Cohen, 1995) inmates. The inventory showed the high
validity of self-reports for both sexual (=.89) and physical (=.88) abuse (Cohen,
1995).
2. Hebrew version of IES - Impact of Event Scale (Horowitz, Wilner & Alvarez.
1979).
This questionnaire measures three types of PTSD: intrusion, avoidance and
emotional avoidance (Joseph, Williams, Yule and Walker, 1992; Myslobodsky et al,
1995). The Hebrew version showed a reliability of =.93, and was found to fit various
population groups (Solomon, 1988). 
Procedure
After signing a written consent form, subjects were individually interviewed.
This method was preferred because many of the prisoners have poor writing skills.
The interviewer explained the purpose of the study, pointing out that participation had
no legal implications, did not entitle the subject to any privileges, and did not
constitute an admission of guilt for the alleged offense. Each prisoner signed an
agreement to take part in the research.
Results
The findings indicated significant age differences between the three groups of
offenders: (F(2,110)=3.20; p<.05) where SO (sex offenders) were older than the
other two groups, with a mean age of 40, while the mean age of VO (violent
offenders) was 38, and that of MO (miscellaneous offenders) was 33. No
5
differences were found between the research groups regarding marital status,
education level and country of origin. However, a significant difference was found
in the number of previous arrests (26=47.5; p<.000): only 25.6% of the SO had
previous arrests compared to 58.6% of VO and 87.2% of MO. In addition, most of
the SO served in the army. Only 21.3% SO did not serve compared to 27.6% of the
VO and 61.7% of the MO (24 = 7.1; p<.05). This finding may suggest that SO do
not, in essence. belong to the criminal sub-culture.
Of the 123 prisoners, 15.4% reported sexual abuse (SV), and 36.6% indicated
physical abuse (PV). No direct correlation was found between victimization type
and type of offence. Table 1 presents the distribution of these two variables.
---------------------------------------Insert table 1 about here
---------------------------------------A three-way MANOVA was used to calculate the relationship between type of
offence, type of victimization and the severity of PTSD symptoms. The following
effects were found to be significant: type of victimization (F(2,105)=5.15; p=. 007),
where PTSD symptoms were more severe in SV and PV than in NV; the type of
PTSD was significantly different between the three groups (F(2,210)=24.8; p=.000).
Among SV, SO suffer from the most severe cognitive avoidance and emotional
avoidance, and they suffer less from intrusive thoughts compared to other groups of
offenders.
Chi Square was used to test the relationship between drug usage and the type
of offences, which was found to be significant (22 = 22.820; p<.01). More than half
of the SO did not use drugs (57.4%), while most of the MO (88.9%) and VO (72.4%)
did use drugs.
A three -way MANOVA was used to assert the relation between type of
offence, type of victimization and drug use. The following main effect was found to
be significant: drug usage (F(1,131)=15.131; p=.000). Among SV, most (86.7%) SV
who were SO did not use drugs, while most (66.7%) SV who were using drugs were
not SO, (22 = 21.16; p<.01). Among PV, a similar trend was found. More than half
(55.6%) of those who were not using drugs were VO, while most (80%) of FV who
were using drugs were not VO, (22 = 5.78; p=.056).
6
We hypothesized that PTSD symptoms would facilitate the SL. In other words,
past victimization would predict the type of criminal activity, while drug abuse would
indicate the GL. Specifically, among drug abusers, no similarity between type of
victimization and type of criminal activity would exist. In order to check our
hypothesis, we used the Analysis of Moment Structure (AMOS) (Arbuckle, 1997).
AMOS (Analysis of Moment Structure) uses structural equation modeling (SEM), a
form of multivariate data analysis that can test for accuracy between the data obtained
in the research and the researcher’s specified model. AMOS calculates maximum
likelihood estimates from the covariance matrix and several fit indexes in order to
evaluate the reliability of the connection between the data and the specified model.
Since AMOS calculates several fit measures, the choice of fit is arbitrary.
However, Hu & Bentler (1999) recommended, for a sample size of 200 or less, that
the following fit measures be used:
χ2
>0.5; CMIN/DF<3 (chi square divided by
degrees of freedom); RMR (Root mean residuals) <0.6; TLI (Tucker-Lewis
index)≥0.95; CFI (Comparative fit index)≥0.95; RMSEA (Root mean square error of
approximation)≤0.06.
Specifying the Model
The model we derived is presented in fig.1. Our indices, presented in table 2,
indicate a good fit between the model and the data. In addition, according to our
hypothesis, drug abuse had a negative effect on crime type. In other words, the higher
the level of substance abuse, the less similarity existed between type of victimization
and type of crime committed.
---------------------------------------Insert table 2 about here
------------------------------------------------------------------------------Insert figure 1 about here
---------------------------------------The model indicates three or four main paths:
1. PTSD developmental path: Trauma (type of victimization) initiates the development of
PTSD, starting with intrusive thoughts leading to cognitive avoidance, and then to
7
emotional avoidance
2.
Two paths indicating GL:
a. Trauma (type of victimization) initiates intrusive thoughts leading to drug abuse and
to criminal activity
b. Trauma (type of victimization) initiates intrusive thoughts leading to cognitive
avoidance ,drug abuse and to criminal activity
3. A SL path: Trauma (type of victimization) initiates intrusive thoughts leading to
cognitive avoidance and to criminal activity, indicating that PTSD has a sequential
development as described in the literature
Discussion
The literature is rich with phenomenological and statistical reports stating
that interesting and intriguing phenomena exist, such as the transformation of
victims into victimizers. However, only a few attempts have been made to explain
these relationships. Moreover, contradictory findings in the literature regarding the
nature of these connections appear (specific link, i.e. victimizations leading to same
offence as victimization or general link, in which any type of victimization leads to
any type of criminal activity). Explanations for this contradiction have focused
largely on poor research techniques or on differences in data sources (Cohen,
1995). Attempts to account for these phenomena theoretically are rare, if they can
indeed be found.
The present investigation directs attention to the role played by
victimization early in life on subsequent PTSD symptoms leading either to drug
abuse and atypical criminal behavior, or directly to criminal behavior similar to that
of the victimization type.
In employing causal modeling, the objective is to hypothesize a model
linking initial variables to indicator variables, where applicable, and a structural
model displaying the functional relations between the variables. The viability of
this model is then assessed by determining the degree to which the indicator
variables can be accounted for in the predicted model. In order to test our model, a
confirmatory factor analysis using AMOS software (Arbuckle, 1997) was
conducted. As shown in illustration no. 1, the analysis introduced two variables as
8
major indicators, drug abuse and PTSD, or to be more exact, intrusive thoughts and
cognitive avoidance.
Whereas drug abuse indicated non-specific connections between past
(childhood/adolescent) types of victimization and types of adult criminal behavior,
PTSD symptoms, such as intrusive thoughts and cognitive avoidance, led to
specific types of criminal behavior (the same as that of the victimization type).
Thus, sexual victims who develop cognitive avoidance tend to become sexual
offenders as adults, while those who develop drug dependency tend to exhibit nonspecific criminal behavior.
This supposition is supported by our finding that most of the sex offenders
who were sexually victimized are not using drugs. In contrast, sexual victims who
are not sex offenders usually use drugs. A similar trend was found among FV.
An abused child learns to be unaware of feelings or unable to express them
because feelings of pain are so strong that they become intolerable. As adults, some
of the victims are able to cope with past trauma, adjust and lead normative lives.
Others learn to survive the abuse by negating their true feelings (Bat Or, 1992).
Such negation can be achieved through addiction to a substance or substances, such
as alcohol, drugs, or food (Teusch, 2001; Bat Or, 1992). However, there is a
connection between drug use and crime. Drugs are very expensive. Moreover,
heavy drug users are usually unable to maintain employment. As a result, many of
them commit crimes in order to finance their drugs (Bennett & Holloway, 2005;
Brochu, 2001). It is easy to deduce that these criminal acts are driven by the need
to purchase drugs, or by the drug-usage lifestyle. Thus, such a pattern comprises
the aforementioned General Link.
The Direct Link, as the finding suggests, has no connection to drug usage.
There is a direct link through intrusive thoughts and cognitive avoidance from
victimization type to type of criminal activity. Children who experience physical
violence or sexual abuse might perceive it as common or acceptable behavior, as a
way to relieve anxiety, or to express anger and/or rage (Barnard et al, 1992). For
these children, the normal response to crisis situations in adulthood may be violent
behavior. This pattern is facilitated by cognitive avoidance that negates the feelings
of compassion to the sufferings caused by their actions.
In conclusion, while it seems strange that one who has suffered abuse will
impose suffering upon others, our findings suggest that the path from victimization
9
to criminal behavior is facilitated by PTSD and drug usage. The findings also
clarify the distinctions between GL and SL.
Acknowledgments
This paper is based on an MA thesis submitted by the second author, and
supervised by the first.
References
Addad, M. (1989) The formation of a criminal. Ramat-Gan: Ur Am (Hebrew).
Cohen, D.A. (1995). The relation between sexual offending and childhood
victimization. An unpublished MA thesis, Department of Criminology, BarIlan University, Rama-Gan.
American Psychiatric Association. (1994). Diagnostic and Statistical Manual of
Mental Disorders, 4d ed. Washington, D.C. American Psychiatric Press.
(DSM IV).
Arbuckle, J.L. (1997). Amos user’s guide: version 3.6. Chicago: SmallWaters Corp.
Bandura, A. (1973). Aggression: A Social Learning Analysis. N.J.: Prentice Hall.
Bandura, A. (1977). A Social Learning Theory. New-York.: Prentice H all.
Barnard, G.W., Hankins,G.C., & Robbins, L. (1992). Prior life trauma, post
traumatic symptoms, sexual disorder, and character traits in sex offenders:
An exploratory study. Journal of Traumatic Stress, 5, 393 - 420
Bat-Or, R. 1992. Seven Steps to Healing, A Healing Model of adults who were
abused as children. in: Ben-David S., and Kirchhoff G.F. (Eds):
International Faces of Victimology. Munchengaldbach: World Society of
Victimology Press: 365-382
Ben - David, S.( 1991). Childhood and adolescent victimization of female
prisoner. In : Kury H. International Research on Victimology. Freiburg,
(Max Plank Institute), pp 459 - 478.
Ben - David, S. (1994) “Why they don’t remember? Crime as an immune device
against traumatic intrusive recollections” in: Kirchhoff G.F., & Kosovski. E
11
(eds). International Debates of Victimology. Monchengladbach: MSV
Publishing
Ben-David, S., Alek, M. and Silfen, P. (2002). Victimization, trauma, and criminal
activity among prisoners in Israel. In: Addad, M. and Wolf, Y. (Eds.) Crime
and social deviance. (pp:107-124). Ramat-Gan: Bar-Ilan University Press
(Hebrew with English abstract)
Bennett, T., Holloway, K. (2005). The Association between multiple drug misuse
and crime. International journal of offender therapy and comparative
criminology, 49: 63-81
Bergen, H.A., Martin, G.,Richardson, A.S., Allison, S., Roeger, L. (2004). Sexual
abuse, antisocial behavior and substance use : gender differences in young
community adolescents. Australian and New Zealand journal of psychiatry,
38 :34-41
Brochu, S. (2001). The relationship between drugs and crime. Quebec, Canada:
University de Montreal. Retrieved March 3, 2005 from
http://www.parl.gc.ca/37/1/parlbus/commbus/senate/Com-e/illee/presentation-e/brochu-e.htm
Clarke, J., Stein, M.D., Sobota, M., Marisi, M., Hanna L. 1999. Archives of
International Medicine, 159:1920-1924
Curtis, G.C. (1963). Violence breeds violence - perhaps. American Journal of
Psychiatry, 120, 386.
Everly, G.S.(1995). An integrative two factor model of post traumatic stress. In:
Everly, G.S., & Lating, J.M.(eds). Psychotraumatology, Key Papers and
Core Concepts in Post Traumatic Stress. 27 – 48.
Finkelhor, D. (1988). The trauma of child sexual abuse: two models. In : Wayte,
G.E. & Powell, G.J. (eds.) Lasting Effects of Child Sexual Abuse. London:
Sage.
Freud, A. (1966). “The Ego and the mechanisms of defense. New York:
International Universities press
Foa, E.B., Riggs, D.S., & Gershuny, B.A. (1995).Arousal, numbing, and intrusion:
Symptom structure of PTSD following assault. American Journal of
Psychiatry. 152:1, 116 – 120.
11
Graham K.R. 1996. The childhood victimization of sex offenders: An
underestimated issue. International Journal of Offender Therapy and
Comparative Criminology, 40:3, 192-203
Haapasalo J., & Pokela E. 1999. Child –rearing and child abuse antecedents of
criminality. Aggression and Violent Behavior, 4, 107-127
Horowitz, K., Weine, S., Jekel, J.(1995). PTSD symptoms in urban adolescent
girls: Compounded community trauma. Journal of Academy Children and
Adolescents Psychiatry, 34:10, 1313 – 1361.
Horowitz M., Wilner N., & Alvarez. W. (1979). Impact of Event Scale: A measure
of subjective stress. Psychosomatic Medicine, 14, 3: 209 - 218.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance
structure analysis: conventional criteria versus new alternatives. Structural
Equation Modeling, 6, 1-55
Joseph, S.A., Williams, R., Yule W., Walker, A.( 1992). Factor analysis of the
impact of events scale with survivors of the two disasters at sea. Personality
and Individual Differences, 13:6, 693 – 697.
Laws, D. R., & Marshall, W. L. (1990). A conditioning theory of the etiology and
maintenance of deviant sexual preference and behavior. In: W.L. Marshall,
D.T. Laws, & H. E. Barbaree (Eds.), Handbook of sexual assault: Issues,
theory, and treatment of the offender (pp.209-229). New York: Plenum
Lipton, R.J.(1988). Understanding the traumatizes self: Imagery, symbolization
and transformation. In : J.P. Wilson, Z., Harel., & B. Kahan. (eds). Human
Adaptation to Extreme Stress: From the Holocaust to Vietnam. New York:
Plenum.
Marshall, W.K., & Barbaree, H.E. (1990). An integrated theory of the etiology of
sexual offending, In: W.L. Marshall, D.T. Laws, & H. E. Barbaree (Eds.),
Handbook of sexual assault: Issues, theory, and treatment of the offender
(pp.257-275). New York: Plenum
Maxfield, M.G. & widom C.S. 1996. The Cycle of violence – Revisited 6 years
later. Archives of pediatrics & adolescent medicine. 150: 390-395
Myslobodsky, M.S., Glicksohn, J., Singer, J., Stern, M., Bar Ziv, J., Friedland, N.,
& Bleich, A. (1995). Changes of brain anatomy in patients with
posttraumatic stress disorder: A pilot magnetic resonance imaging study.
Psychiatry Research 58: 259 – 264.
12
Nachshon, I. (1988). Hemisphere function in violent offenders. In: T.E. Moffit,
S.A. Mednick & S.A. Stack (eds). Biological Contributions to Crime
Causation ,(pp. 55 – 67). Dordrecht, Netherlands: Nijhoff.
Rowan, A.B., & Foy, D.W. (1993). Post traumatic stress disorder in sexual abuse
survivors: A literature review. Journal of Traumatic Stress, 6, 3 - 20.
Saladin, M.E., Drobes, D.J., Coffey, S.F., Dansky, B.S., Brady, K.R., Kilpatrick,
D.G. (2003). PTSD symptoms severity as a predictor of cue-elicited drug
craving in victims of violent crime. Addictive Behavior, 28, 1611-1629.
Smallbone, S.P., & MacCabe, B-A. 2003. Childhood attachment, Childhood
Sexual Abuse, and Onset of Mastrubation among adult sexual offenders.
Sexual Abuse: A journal of Research and Treatment, 15, 1:9
Solomon Z. (1988). Convergent validity of post traumatic stress disorder
(P.T.S.D) diagnosis: Self - report and clinical assessment. Israel Journal of
Psychiatry and Related Sciences , vol 25: 46 - 55.
Spatz-Widom, C. (1992). The cycle of violence. Research in brief no. 5, National
Institute of Justice, US Department of Justice.
Spatz-Widom, C. (2000). Childhood victimization: Early adversity later
psychopathology. National Institute of Justice Journal, Jaguar: 3-9
Stott, D.H. (1950). Delinquency and Human Nature. The Carnegien United
Kingdom Trust, Dunfermline
Teusch, R. (2001). Substance abuse as a symptom of childhood sexual abuse.
Psychiatry services, 52:1530-1532
White HR, Spatz-Widom, C. (2003). Inmate partner violence among abused and
neglected children in young adulthood: The mediating effects of early
aggression, antisocial personality, hostility and alcohol problems. Aggressive
Behavior, 29:332-345
Yoshinaga, C., Kadomoto, I., Onati, T., Sasaki, T., & Kato, N. (2004). Prevalence
of post-traumatic stress disorder in incarcerated juvenile delinquents in
Japan. Psychiatry and clinical neurosciences, 58: 383-388
Zhang, L., Welte, J.W., & Wieczorek, W.F. (2001). Deviant lifestyle and crime
victimization. Journal of crime justice, 29: 133-143
13
Table1. Distribution of Type of Offence among Types of Past Victimization
Types of past victimization
Types of offence
Sexual abuse
Physical
No abuse
Total
(SV)
abuse (PV)
(NV)
n
Sex offenders (SO)
42.1
35.6
39.0
47 (100)
Violent offenders (VO)
42.1
31.0
20.3
29 (100)
Miscellaneous offenders MO 15.8
44.4
40.7
47 (100)
Total N (%)
45 (36.6)
59 (48.0)
123 (100)
19 (15.4)
(%)
14
Table 2. Estimated fit Measures for the Model
χ2
CMIN/DF
RMR
TLI
CFI
RMSEA
9.54
1.19
0.03
0.98
0.98
0.039
note: χ2 (chi-square); CMIN/DF (chi square divided by degrees of freedom ); RMR (Root mean
residuals); TLI (Tucker-Lewis index); CFI (Comparative fit index); RMSEA (Root mean square
error of approximation)
15
Figure1. Estimated Relationship between the Variables in the Model
VICTIMIZ
e2
e1
.08
EMOTIONA
e3
-.30
.34
.28
COGNITIV
.09
.59
-.13
.60
INTRUSIV
.12
.36
DRUGS
e5
.12
-.29
CRIME
e4
16